Author: stuart taylor
Date: 02:42:46 05/08/02
Go up one level in this thread
On May 08, 2002 at 04:52:20, Thorsten Czub wrote: >On May 07, 2002 at 19:56:28, John Reynolds wrote: > >>I have seen some negative remarks about hiarcs, is it as much a disappointment >>as some say? > > >i have results with 2 autoplayer pairs. >one 400 mhz autoplayer pair. one 1200 mhz autoplayer pair. > >i do always play 40/120 games. > > >in this setting, hiarcs is as strong as fritz, shredder , junior, ... THAT'S something QUITE different to the impression people are giving, if its true. Does that mean that at half an hour per move, it could be consistently STRONGER than the other programs? If so, I wouldn't bother upgrading from Hiarcs 8.0 for another 5-10 years, as it will only get better and better with faster hardware, without further intervention, at Blitz too. Also, perhaps It's the strongest available for analysis? And if it's the most interesting program, as you say, then noting can be better, and I've found what I wanted. I mean, I wanted the strongest, and this would be it. And I'm prepared from the label "strongest" to mean when you leave it thinking for half an hour or more. What could be better? Is this all true? Can that be tested? S.Taylor >but the best thing is: > >the games are the most exciting things i have ever seen. > >when i see others post their results, its often blitz games. >in blitz games i can imagine that hiarcs has problems. >hiarcs is IMO tuned for 1'-3' per move games. > >This was 1993 the case with hiarcs2, and it is today the case with hiarcs8. > >the best thing in the games is the different opinion about the game-position. >hiarcs feels itself in advantage. fritz itself. and both try to "convince" >each other. > >thats very very interesting. a program that sees the position different. > >i doubt that the short games of the others give a complete image about hiarcs >strength. > >its a difference if you play blitz games or tournament games. >IMO in blitz games, especially when it comes near to the time control, >the speed of the program is very important. > >and here fritz and junior have clear advantages. > >in 40/120 games hiarcs can show what it knows about chess, and that it plays >like a human beeing. > >IMO one can use hiarcs8 very good for analysis of email-chess games. >hiarcs7.32 was brilliant in that field, and i think hiarcs8 is even better. > > >Since 40/120 games take more time than 60' blitz, i cannot post series of games. >But when i post ONE game (e.g. in the moment it is played out) >people jump on the PGN posting and give shit comments. so i do not post my >games anymore. may the ones posting fast games succeed with their stupid >shallow and stupid criticicm far from computerchess. 34-64 bean counting is >indeed something i am not interested in. > >Look. >in the moment e.g. hiarcs8 says +0.56 whikle fritz7.0.0.7 evaluates +0.31 >for itself too. > >the position: > >w K g1 Qd2 Rc1c5 Be2b2 a3d4e3g2h2 >b K g8 Qd6 Rb8f8 Be6 Ne7 a7e4f7g7h7 > >hiarcs is white. > >the game went out of book arround move 13. another advantage of hiarcs. >it produces games that begin very early with engine computing. no long lines >of idiotic opening books. > >How would YOU evaluate the chances of white and black in this position ? > >This is what i like the most. I can guess with the computers. see how they >do it. > >i can get a 3rd machine and analyse my ideas about the game-position. > >such a game is computerchess at its best for me. >it could have happen on world championships. or in your living room now!
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.